Biostatistics Shared Service The goal of the Biostatistics Shared Service is to provide basic, translational, clinical, and epidemiology investigators at the University of Maryland Marlene and Stewart Greenebaum Cancer Center (UMGCC) with access to high quality statistical science. Functions of the shared service include designing, reviewing, and monitoring protocols;statistical analysis, planning, and reporting;designing and analyzing high dimensional data, including microarray gene expression, single nucleotide polymorphisms (SNPs), epigenetics, proteomics;preparing grant applications and manuscripts;providing biostatistics training and advice on clinical data management;and database computing. The Biostatistics Shared Service provides investigators at UMGCC with advanced statistical methods and expertise developed under other grants funded by the National Cancer Institute, such as design and analysis of combination studies, analysis of microarray and cancer epidemiology data, and SNP data mining. Statisticians in this shared service collaborate with investigators in all three cancer programs at UMGCC: Experimental Therapeutics, Hormone Responsive Cancers, and Molecular and Structural Biology. The shared service has access to excellent computing resources and statistical software packages (e.g., SAS, StatXact, Splus), database management software (e.g., Access, Oncore[unreadable]), and specialized statistical software necessary for efficient and effective support of UMGCC's programs. From July 2006 to June 2007, this shared service supported 128 projects from 60 investigators and 42 grant applications. Nearly 55 percent of total effort was spent on fund.ed cancer projects. For this grant, we request 0.50 full-time equivalent (FTE) of biostatistics and bioinformatics support and 0.25 FTE of administrative and database assistance to provide biostatistics and bioinformatics support to each cancer program at UMGCC.